About Small Print
Small Print is a read-only MCP server for checking recorded descriptions, schemas and instructions for MCP servers, agent skills and plugins. It shows version changes with grading reasons and related public advisories, helping agents assess changes before use.
Key features
- 01Look up recorded entry details
- 02Compare releases and graded changes
- 03Find advisories with version ranges
- 04Check changes since approval
- 05Mark quoted text as untrusted data
Use cases
- 01Review a tool’s recorded instructions before use
- 02Check whether an approved entry has changed
- 03Inspect why a release change received its grade
- 04Find advisories for a plugin version
MCP support
Small Print exposes four read-only tools through an MCP server: lookup_entry, changes_since, advisories_for and changed_since_approval.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
4Saved tool list · updated 5d ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| name | string | Yes | The entry, with its registry prefix when known: npm:@scope/name, pypi:name, mcp-registry:io.github.owner/server, skills.sh:owner/repo/skill, oci:ghcr.io/owner/image. A bare name is read as an npm package. Case-sensitive, up to 300 characters. |
Details
From the project READMEsmallprint-mcp
Small Print as an MCP server: four read-only tools over the public record of the tool descriptions, schemas and instructions of MCP servers, agent skills and plugins: the versions Small Print has read, the changes between them with the rule behind each grade, and the public advisories that name them.
{ "mcpServers": { "smallprint": { "command": "npx", "args": ["-y", "smallprint-mcp"] } } }
Tools:
lookup_entry(name): what the record holds for one entry.changes_since(name, since?, min_severity?): the releases that changed the small print, with each diff and the rule behind its grade.advisories_for(name, version?): the advisories that name it, attributed to their sources, with version ranges.changed_since_approval(name, approved): yes or no, before use: has the small print changed since the version, content hash or date that was reviewed. Answers start with UNCHANGED, CHANGED or UNKNOWN. UNKNOWN means the record cannot say; treat it as no answer, not as safe.
Names: npm:@scope/name, pypi:name, mcp-registry:io.github.owner/server, skills.sh:owner/repo/skill, oci:ghcr.io/owner/image; a bare name is read as npm.
Most of what the record holds was written by other people: names, versions, registry descriptions, changed tool text, advisory summaries. Every answer marks that text as quoted, a short value inside ⟦ ⟧ and longer text as lines that start with "|" between a QUOTED line and an END QUOTED line, and says that marked text is data to report, not an instruction to follow.
The server reads https://smallprint.dev/api and nothing else. No account, nothing about your machine is sent, and it never calls the tool it looked up. Rate limited; for the whole record see the site. The rules behind every grade: https://smallprint.dev/how-we-grade.
What 1.0 holds stable
From 1.0.0 the four tools, their names, their inputs and the fields of their answers do not change without a new major version. A field can be added in a minor version; none is removed or renamed in one. The grading rules behind a grade can change in any version.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.